1 / 25

Spatial Clustering in Juvenile Delinquency and Recidivism in Philadelphia, Pennsylvania

Association of American Geographers April 2007. Spatial Clustering in Juvenile Delinquency and Recidivism in Philadelphia, Pennsylvania. Jeremy Mennis, Phil Harris, Zoran Obradovic, Alan Izenman, Heidi Grunwald, Yilian Qin, Joseph Jupin, Brian Lockwood Temple University. A Question.

gusty
Download Presentation

Spatial Clustering in Juvenile Delinquency and Recidivism in Philadelphia, Pennsylvania

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Association of American Geographers April 2007 Spatial Clustering in Juvenile Delinquency and Recidivism in Philadelphia, Pennsylvania Jeremy Mennis, Phil Harris, Zoran Obradovic, Alan Izenman, Heidi Grunwald, Yilian Qin, Joseph Jupin, Brian Lockwood Temple University

  2. A Question What happens when a criminologist, a geographer, a statistician, a computer scientist, and a higher education administrator get together to do interdisciplinary research on juvenile delinquency?

  3. Observations… • Adolescent development and behavior can be supported and hampered by environmental forces. • Aftercare services must address youths’ developmental needs, which may be aggravated by external forces that compete with program effects.

  4. Research Objectives To develop, apply, and evaluate improved techniques to investigate the simultaneous effects of individual, program, and neighborhood forces in preventing juvenile recidivism. A case study focusing on adjudicated juvenile delinquents assigned to court-ordered programs by the Family Court of Philadelphia, Pennsylvania, during the years 1996 to 2003.

  5. Program Development and Evaluation System (ProDES)11,659 male cases

  6. Cases and Juveniles by Tract

  7. Number of Cases Per 100 Juveniles by Tract

  8. Recidivating Cases

  9. Recidivism Rate by Tract

  10. Combinations of Classed Delinquency and Recidivism

  11. Cases: Recidivating and Non-Recidivating

  12. Modeling Recidivism Rate as a Surface For each case, rate is calculated over 100 nearest neighbors. Calculation accounts for major barriers such as rivers and parks. Each case location is considered a sample of the surface and entered into Getis Gi*.

  13. Clustering of Any Recidivism

  14. Clustering of Person Offense Recidivism

  15. Clustering of Recidivism with Removal from the Community

  16. Comparison of Clustering of Outcome Variables Person Offense Recidivism Removal from the Community Any Recidivism

  17. Onward! • Completed data cleaning, dimensionality reduction. • Integrating data on programs, socio-economic status, neighborhood collective efficacy, and crime • Investigate spatial non-stationarity in models of the target variables. • Evaluate HLM and spatial econometric modeling for predicting target variables.

  18. Alan Phil Zoran Jeremy Heidi Enjoy San Francisco! This research is supported by a grant from the National Institute of Justice. Jeremy Mennis jmennis@temple.edu astro.temple.edu/~jmennis

  19. Program Design Inventory (PDI)43 programs for boys

  20. Recidivism by Program

  21. Recidivism by Program

  22. Philadelphia Health Management Corporation (PHMC)45 neighbor-hoods

  23. U.S. Burueau of the Census 365 tracts

  24. Challenges • Data Integration • High-Dimensional and Noisy Data • Spatial Dependency and Heterogeneity

  25. Research Design • Data Preparation • Data Exploration • Model Construction, Evaluation, and Interpretation

More Related